this post was submitted on 19 Jun 2026
714 points (98.1% liked)
Technology
85670 readers
3404 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related news or articles.
- Be excellent to each other!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
- Check for duplicates before posting, duplicates may be removed
- Accounts 7 days and younger will have their posts automatically removed.
Approved Bots
founded 3 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
It isn't the details or severity of the break that matters.
It's that the quality control process is SUPPOSED to catch that, and whatever sorry excuse for a process they're using now ALLOWED a break that was obvious, visible, and repeatable, inside a critical, core function of the operating system, to make it to the end users, something that should trigger as an immediate, flashing warning light. That means the entire quality control process at the very least is SEVERELY compromised and unreliable, and there could very easily be MUCH more severe vulnerabilities and bugs hiding underneath that AREN'T immediately visible. To anyone who has done any professional development for non-disposable code bases, this isn't a whisper of a problem - it's an air horn.
AI found the exploits, and they clearly used AI to fix the exploits.... That about as far as the QC conversation went
Developer written unit tests likely wouldn’t even catch that bug with the recycling bin, because it doesn’t even matter what the text says when it’s being deleted. It’s not a breaking bug. It wouldn’t hold up a release. It might have even been found in QA and might have a super low priority ticket to fix it because again, it’s non breaking and doesn’t affect anything in any way.
You don’t understand how software dev QA works, clearly.
Yeah. I've only been a veteran of the process for 20 years - I don't know a thing about what I'm talking about.
"non-breaking" is a meaningless distinction. What you're REFERRING to is a "cosmetic" bug, and "cosmetic", depending on the software and the shop, does NOT mean "acceptable for release", and FURTHERMORE, this is not a new bug that's been filed as low priority or will-not-fix, but a REGRESSION because it DID work in the past, which means it's DOUBLE damning.
I'm not interested in waving around credentials about who knows more about software development - if you work in a shop that doesn't care about quality, that's between you and the shop. But if you want to claim that someone at Microsoft said "Yeah, it doesn't correctly reflect the filename, a critical check to ensure users don't accidentally delete the wrong file, which is something that's worked for 30 years" and then signed off on that, instead of the MUCH more likely explanation that NOBODY is looking at ANY of this crap with the detail they should be, I'm afraid I'm going to have to laugh.
To give you an idea, in the XBox division, a division of Microsoft, this would be considered a compliance failure that prevents a game from going gold for launch on the platform, and if caught would cost the developer thousands and weeks to fix before the game could go live because it would necessitate starting the final step of the certification process over again, because EVERY SINGLE TEST has to be run again to ensure JUST this kind of regression doesn't resurface.
But no - the same company would claim it's totally acceptable for the operating system that runs bank software because it's "non-breaking".
Non-breaking isn’t meaningless - it means it doesn’t break anything. It means it doesn’t affect users, and thus it’s not going to be high priority or be enough to block a release.